Primary dystonia, which often begins in late childhood or adolescence, has traditionally been attributed to basal ganglia dysfunction, but no specific histopathological lesions of these structures are evident on postmortem examination. In fact, the challenge of primary dystonia is reflected in its clinical definition: the presence of involuntary, sustained muscle contractions in the absence of identifiable brain lesions. This lack of clear neuropathology has this illness difficult to understand and treat. Our recent work supports a new paradigm for understanding dystonia as a neurodevelopmental circuit disorder. Taking advantage of the partial penetrance of inherited dystonia and the fact that dystonic movements disappear during sleep, we employed new magnetic resonance diffusion tensor imaging (DTI) methods in manifesting and non- manifesting mutation carriers to examine motor circuit activity. We made the surprising discovery that both manifesting and non-manifesting mutation carriers exhibit disruptions of the cerebello-thalamo-cortical tract, and that non-manifesting individuals show additional, distal disruptions in the thalamo-cortical projection pathways. This distal defect is clinically protective: it blocks transmission of the aberrant cerebellar output to the motor cortex. Our work has given rise to a new model for the motor circuitry dysfunction underlying dystonia, which we will test in this proposal as we seek to identify the changes in fiber tract integrity (microstructure) and neural activation responses (function) that regulate clinical penetrance, account for phenotypic differences, and modulate treatment response in primary dystonia. In Specific Aim 1 we will characterize motor circuit abnormalities in hereditary primary dystonia by examining structure-function relationships in manifesting and non-manifesting carriers of the DYT1 and DYT6 mutations. We will assess pathway microstructure (using DTI) and circuit function using H215O PET to measure cerebral blood flow during task performance and in the rest state and fMRI to localize task-related neural activation responses. In Specific Aim 2 we will identify circuit abnormalities in sporadic dystonia, conducting DTI/tractography studies and brain activation experiments and comparing the results to those obtained in patients with hereditary forms of the disease. Finally, in Specific Aim 3, we seek to understand how deep brain stimulation (DBS) works in dystonia when it does work, and identify predictors of patient response. DBS can be effective in treating severe primary dystonia, but not all patients benefit equally, and symptoms can re-emerge after an initial period of abatement. Subjects will undergo preoperative imaging and then participate in a series of studies at multiple postoperative time points following the start of chronic stimulation. The resulting scan data will be used to: (a) measure serial changes in network activity as the treatment response develops; (b) identify patterns of microstructural change at baseline that correlate with clinical treatment response; and (c) develop quantitative preoperative imaging descriptors to predict treatment response in individual subjects.